(FORTUNE Magazine) – Neither East nor West excels at sustained innovation, according to esteemed Japanese professor Hirotaka Takeuchi. Winning at the subtle game he calls "knowledge creation" means integrating apparently incompatible ways of thinking.

Takeuchi, 49, ranks among the intellectual leaders of the younger, globally minded generation that is coming to power in Japan. A professor of marketing and international business at Japan's Hitotsubashi University, Takeuchi is teaching an advanced management course at the Harvard business school this spring.

Takeuchi argues that an organization's ability to sustain innovation--and thus revenue growth--depends on creating and spreading knowledge among middle managers. Training isn't enough, and neither is individual genius. How to get the process started and keep it going is the subject of The Knowledge-Creating Company: How Japanese Companies Create the Dynamics of Innovation (Oxford University Press, $25), written with his mentor and colleague, Ikujiro Nonaka. The book draws lessons from a study of Japanese companies with outstanding records in product development.

What does "knowledge creation" mean?

In a word, innovation. We all keep hearing about knowledge workers--but where does the knowledge come from? Our concern was to determine how organizations innovate, not just once or twice but systematically and continuously. Our starting point was that the innovation process in Japan is different from that in the West. And our conclusion is that knowledge creation has nothing to do with Japanese management--it's universal.

Are the cultural differences important?

Yes, because we have to learn from each other. Western companies have a major strength: One individual can quickly bring a big innovation to market. But in a company like Microsoft, you wonder where the great ideas will come from after Bill Gates. Japanese companies rely on innovations from groups of ordinary people. They depend on middle managers' trying to push everyone in the team up to a higher level of shared understanding. That takes longer, and it's less spectacular, but a group's knowledge base can become a major competitive advantage.

The real excitement comes when you combine the two approaches. We saw it when Caterpillar and Mitsubishi jointly developed a hydraulic shovel. For Mitsubishi, the key lesson was how good the Caterpillar guys were at putting everything they knew into a manual. At first the Caterpillar people thought the chorei--the ten-minute staff meeting that started every day--was for the birds. But gradually they realized that getting people together every morning is a major strength of this system.

How does human behavior affect innovation?

We learned that knowledge creation depends on the interaction of two types of knowledge, and that each culture is most comfortable with only one. In the U.S., people are very comfortable with explicit knowledge, the sort of information that can be verbalized, written down in documents, put into computers, and readily communicated. The Japanese don't do that very well. They're best at tapping tacit knowledge, which comes from personal experience and is usually difficult to express. It's what you feel in your gut when you have a hunch or an inspiration. It's not purely rational. Tacit knowledge is connected to emotions, beliefs--it's more a bodily knowledge than something in the mind. Americans aren't as comfortable with that.

Creating knowledge requires a convergence of tacit and explicit knowledge. There has to be a spiral from one type of knowledge to the other, and back again.

How exactly does tacit knowledge apply to competition in the marketplace?

Let me give you an example. In Japan, Matsushita Electric used to be known as maneshita, which means "copycat." Big and successful but not an innovator. That changed dramatically with the introduction of the Home Bakery, the first automatic breadmaker. A software engineer, a woman named Tanaka, recognized that with Westernization, the time had come for a breadmaker in Japan. But she knew almost nothing about baking. So she apprenticed herself to a master baker. He had all the knowledge in his fingertips, but it was very hard for him to verbalize. After watching him for two or three weeks, she went back to Matsushita to write up a set of specifications for the machine, translating his tacit knowledge into something explicit.

They made a prototype, but the bread tasted terrible. So Tanaka-san brought a group of her peers to observe the baker again. Finally, they realized that what the machine lacked was the twisting motion the baker used when kneading his dough. Incorporating that understanding enabled them to develop a hugely successful product.

The breadmaker changed the corporate culture at Matsushita. People in other divisions said, "Why can't we do that?"

How do Japanese companies get the benefit of employees' tacit knowledge?

They work hard to create a feeling of unity. The influence of Zen makes us believe that the best state of intellectual ability comes when the body and mind become one. The ideal form of communication is known as the "breath of ah-un." That's when one person looks at another and says, "ah," and other guy says, "un," and that's all they need to understand each other. No explicit verbal communication at all.

Here's how it works. Years ago when I was at an ad agency, the boss would invite us to a night out every month or so. We'd get totally drunk and tell him off, airing our complaints. The next morning when we met the boss, we'd make a little bow and say, "Ooooh, I was so drunk, I don't remember anything." It's a ritual that clears the air and promotes understanding without any penalties.

Another approach is to hold an off-site meeting called a gasshuku, or camp-in. Team members camp in at a hotel and exchange ideas. Barriers come down. I take my students off-site to a mountain resort with a hot spring. We'll all be soaking together with our clothes off--it's not coed, by the way--and the students will volunteer to rub my back. That's how we communicate. The experience we share is symbolic of our unity. We're all naked in the same pool together. That leads to a deeper understanding, a feeling of trust.

I wonder if Cincinnati is ready for the hot-tub approach. What can Westerners do to get more systematic about tacit knowledge?

Culturally, there are better ways to do this in the West--maybe skiing together, or going to the Outward Bound school. The key is interacting as equals, finding ways to establish human, emotional contact. Top management has to spend much more time rubbing knees with their key middle managers.

Why the emphasis on middle management?

Innovation doesn't require genius. In the U.S. you hear people saying middle managers are dead weight, but these people play a key role in many of the Japanese companies we interviewed. Our corporate leaders are ideal driven. It's up to the frontline people to put those ideals into practice. When information is shared freely between top and middle management, knowledge spirals up and down the organization.

Most of the successful product development we observed was done by project teams. They all work together in a tiny room for two years. Nonaka and I call it the rugby approach: Everybody is on the field from day one--engineers, production, quality control, sales. Whenever a problem comes up, they all pile in together. That's very different from the relay approach, in which one functional team passes the baton to the next.

Project teams work when goals are very clear and they have the autonomy they need to meet them. And the participants are handpicked by top management. Sharp creates special teams for what it calls "urgent projects." When top management decides something is strategically very important, everyone on the team gets a gold badge. If you work for Sharp, and someone wearing one of those badges asks you for mission impossible, you do it on the spot. Believe me, the people who wear those badges learn a lot.

The kind of institutional knowledge we're talking about is created when knowledge passes from the individual to the team, and then from the team to the organization. So when a project is completed at a company like Sharp, the team members return to the normal hierarchy. That gives them a chance to dump all the knowledge they have accumulated from their experience into the common-knowledge base. The understanding of their colleagues goes up a notch. Gradually, the experience of project teams spreads through the whole organization. That's how you achieve continuity and repeatability.

Layoffs are a fact of life in the U.S. That practice seems antithetical to the building of organizational knowledge.

If you're letting go of people left and right, you're definitely cutting down your knowledge base and reducing trust. But some employees have a richer knowledge base than others, so if a company is forced to lay people off, it's important to have someone around who understands that. In Japan some of the best guys go into human resources. U.S. organizations would be ahead of the game if they at least developed a human resources function that viewed people as strategic resources.

Any other advice?

Americans could be much less rigid about benchmarking. The Japanese approach to product development is loosey-goosey: Top management sets the vision and mission, then lets the team run. The Western approach can be very mechanical and explicit, so the team gets drowned with measurements, measurements, measurements. The point is to achieve the mission.

REPORTER ASSOCIATE Erin M. Davies